Reliable scale-down models combined to high throughput enable DoE studies for faster and improved process optimization

An industrial bacterial upstream process was improved with the implementation of a new production medium. The aim of the work presented here was to revisit process regulations set-ups in order to get a better knowledge of design space and find the best operating conditions. This was done through a methodology based on efficient scale-down model, followed by design of experiments studies which enable both a better knowledge of design space and process optimization. High throughput scale-down models enabled us to investigate the design space in order to find the best operating conditions for maximizing antigen production and minimizing major impurities production. The study was completed in 3 weeks, compared to roughly 4 months with previous scale-down models and methodology. Finally, the process was successfully scaled up back to lab scale (30L) and industrial scale (1000L). This methodology is now used on process development, troubleshooting and optimization. With the development of single-use high throughput scale-down models and statistical software, we can have an extensive understanding of upstream processes within a short timeframe.